[ASC-media] Media release: breakthrough in robot mining

CRCA Media crca-media at starclass.com.au
Mon Apr 5 19:16:00 EST 2004

CRCA Media		04/ 12
April 6, 2004


Australian scientists have taken a critical step towards automation of underground coal mining for efficiency and safety, with the development of a tool to detect, log and identify production stoppages. 

As part of the CSIRO/CRCMining Landmark Project on Longwall Automation, researchers at the CRCMining have developed a computer system for on-line analysis of trend monitoring data to detect and locate coalface equipment faults.

Without effective fault detection, the dream of fully automating longwall coal mining operations and removing miners from the riskiest places will not be achievable, says the Centre's Prof Hal Gurgenci.

"The key to automation is reliability.  From the information we are gathering we are creating models that not only log what goes on - tremendously valuable for the industry - but should soon be able to predict faults and breakdowns before they occur."

"As underground coal mining costs amount to several billion dollars annually , the ability to reduce delays and downtime could save the industry tens of millions of dollars."  

With up to 10,000 variables to be constantly monitored at the coalface the task is a huge one, requiring broadband communication and the latest in complex systems analysis, Prof Gurgenci says.

One outcome of the project is already available for application in the mine. This is automatic processing of data to generate daily or weekly reports on:
o uptime/downtime
o average production delay
o a log of stoppages in each shift and their durations.

"We demonstrated that the particular data-driven techniques adopted in this project enable us to correctly detect and isolate over 90% of the target faults with misclassification rates lower than 20%," he says.

In Australian longwall mines, machine downtime due to planned and breakdown maintenance represents over 35 per cent of the total mine hours on average. With underground mining contributing one third of the coal industry's $12 billion export revenue, reducing stoppages has major implications not only for industry profitability but also for the economy. 

The work is being carried out in two major underground mines in Queensland's Bowen Basin, where equipment-related delays account for over 50% of all lost production time.

Prof Gurgenci says the next step in the research is the development of ways to predict the loads and stresses that lead to breakdowns in key longwall machinery - the shearer, the armoured face conveyor and roof support, in particular, which between them account for a quarter of all downtime.

Besides helping to improve productivity in existing mines, the work is aimed at even greater efficiencies in future, Prof Gurgenci says. "The typical longwall is 250 metres in length but there's already talk of going to 400 metres.  This means we need a much clearer understanding of the load factors if we are to design machines that can reliably withstand them."

He adds that the logging of reliable data on faults will in itself yield significant short-term dividends for the coal industry, enabling engineers to target the most troublesome areas and equipment.  

In the next phase this will then lead to more accurate fault prediction and avoidance.

"All this contributes to the ultimate goal of automating the process of longwall mining, through a combination of robotics and remote control and using advanced geophysics to predict exactly where the coal seam is."

The Australian Coal Association Research Program (ACARP) has designated automation of longwall mining as one of the industry's key strategic challenges.  The CRCMining and CSIRO are partners in the ACARP Landmark Project on Longwall Automation.

"The benefit for the industry will be a potentially higher, more consistent production rate and the removal of face workers from more hazardous and dusty areas, Prof Gurgenci explains.. 

The CSIRO/CRC Landmark project has been running for two years and achieved several important milestones:
"	On-line 3D shearer position information is now routinely available.
"	Wireless Ethernet has been shown to be a viable, robust coal face communications system
"	EtherNet/IP has been adopted as the standard for communications.
"	Condition monitoring analysis suggests the feasibility of implementing an on-line trend and condition monitoring system.
"	Major longwall mining companies are actively supporting the project and are partnering technical developments
"	The benefits of other-industry technology such as INS, thermal imaging, processor control, data communications are being transferred successfully to longwall automation.

"Although the task remains complex, the risks are relatively low as many of these technologies have been proven in other areas," Prof Gurgenci says.

"The focus on productivity and designing the system for coping with exceptional issues will also ensure a lower risk and provide an incentive for progressive uptake of automation technology in Australia's coal mines - making them among the world's first to do this."

The research addresses National Research Priority three - frontier technologies to transform industry.

More information:
Prof. Hal Gurgenci, CRCMining	07 3365 3607 or  0400 761 224
Prof. Mike Hood, CRC Mining		07 3365 5641
Web:  www.crcmining.com.au
Prof. Julian Cribb, CRCA media		0418 639 245

More information about the ASC-media mailing list